| Message ID | 20200826120421.44356-1-guilhem@barpilot.io (mailing list archive) |
|---|---|
| State | RFC, archived |
| Headers | show |
| Series | intel_idle: Add ICL support | expand |
On Wed, Aug 26, 2020 at 2:05 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > Use the same C-states as SKL Why is this change needed? > Signed-off-by: Guilhem Lettron <guilhem@barpilot.io> > --- > drivers/idle/intel_idle.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c > index 8e0fb1a5bdbd..1bb539f09a4f 100644 > --- a/drivers/idle/intel_idle.c > +++ b/drivers/idle/intel_idle.c > @@ -1145,6 +1145,8 @@ static const struct x86_cpu_id intel_idle_ids[] __initconst = { > X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L, &idle_cpu_skl), > X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE, &idle_cpu_skl), > X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_X, &idle_cpu_skx), > + X86_MATCH_INTEL_FAM6_MODEL(ICELAKE, &idle_cpu_skl), > + X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_L, &idle_cpu_skl), > X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X, &idle_cpu_icx), > X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNL, &idle_cpu_knl), > X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNM, &idle_cpu_knl), > -- > 2.27.0 >
On Wed, 26 Aug 2020 at 14:43, Rafael J. Wysocki <rafael@kernel.org> wrote: > > On Wed, Aug 26, 2020 at 2:05 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > > > Use the same C-states as SKL > > Why is this change needed? On my laptop, a Dell XPS 13 7390 2-in-1 with i7-1065G7, ACPI only report "C1_ACPI", "C2_ACPI" and "C3_ACPI". After this patch I have "C1", "C1E", "C6", "C7s", "C8", "C9" and "C10". I tested for some time on idle (without any application launched) and battery consumption drop ~1W/2W. But it can be a measurement problem or different context. So don't hesitate to tell me if I look wrong or if ACPI is a better option for this specific processor family. > > Signed-off-by: Guilhem Lettron <guilhem@barpilot.io> > > --- > > drivers/idle/intel_idle.c | 2 ++ > > 1 file changed, 2 insertions(+) > > > > diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c > > index 8e0fb1a5bdbd..1bb539f09a4f 100644 > > --- a/drivers/idle/intel_idle.c > > +++ b/drivers/idle/intel_idle.c > > @@ -1145,6 +1145,8 @@ static const struct x86_cpu_id intel_idle_ids[] __initconst = { > > X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L, &idle_cpu_skl), > > X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE, &idle_cpu_skl), > > X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_X, &idle_cpu_skx), > > + X86_MATCH_INTEL_FAM6_MODEL(ICELAKE, &idle_cpu_skl), > > + X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_L, &idle_cpu_skl), > > X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X, &idle_cpu_icx), > > X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNL, &idle_cpu_knl), > > X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNM, &idle_cpu_knl), > > -- > > 2.27.0 > >
On Wed, 2020-08-26 at 15:03 +0200, Guilhem Lettron wrote: > On Wed, 26 Aug 2020 at 14:43, Rafael J. Wysocki <rafael@kernel.org> wrote: > > On Wed, Aug 26, 2020 at 2:05 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > > Use the same C-states as SKL > > > > Why is this change needed? > > On my laptop, a Dell XPS 13 7390 2-in-1 with i7-1065G7, ACPI only > report "C1_ACPI", "C2_ACPI" and "C3_ACPI". Did you try to dig into the BIOS menus and check if you can enable more/deeper C-states? Artem.
On Wed, 2020-08-26 at 15:03 +0200, Guilhem Lettron wrote: > On Wed, 26 Aug 2020 at 14:43, Rafael J. Wysocki <rafael@kernel.org> wrote: > > On Wed, Aug 26, 2020 at 2:05 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > > Use the same C-states as SKL > > > > Why is this change needed? > > On my laptop, a Dell XPS 13 7390 2-in-1 with i7-1065G7, ACPI only > report "C1_ACPI", "C2_ACPI" and "C3_ACPI". Also, if you could runt turbostat - we could see which _actual_ HW C- states are used on your system, which Package C-states are reached. Just get a reasonably new turbostat (it is part of the kernel tree, you can compile it yourself) and run it for few seconds (like 'turbostat sleep 10'), get the output (will be a lot of it), and we can check what is actually going on with regards to C-states. Artem.
On Wed, Aug 26, 2020 at 2:59 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > On my laptop, a Dell XPS 13 7390 2-in-1 with i7-1065G7, ACPI only report "C1_ACPI", "C2_ACPI" and "C3_ACPI". Can you list the properties of the idle states in this configuration, that is: $ grep -r '.*' /sys/devices/system/cpu/cpu0/cpuidle/state1/ and analogously for state2 and state3, and send the output of this? On my desktop system the above comment produces the following output: /sys/devices/system/cpu/cpu0/cpuidle/state1/disable:0 /sys/devices/system/cpu/cpu0/cpuidle/state1/above:1022634 /sys/devices/system/cpu/cpu0/cpuidle/state1/time:2556533185 /sys/devices/system/cpu/cpu0/cpuidle/state1/power:0 /sys/devices/system/cpu/cpu0/cpuidle/state1/residency:2 /sys/devices/system/cpu/cpu0/cpuidle/state1/latency:2 /sys/devices/system/cpu/cpu0/cpuidle/state1/usage:17625391 /sys/devices/system/cpu/cpu0/cpuidle/state1/desc:MWAIT 0x00 /sys/devices/system/cpu/cpu0/cpuidle/state1/below:6578613 /sys/devices/system/cpu/cpu0/cpuidle/state1/default_status:enabled /sys/devices/system/cpu/cpu0/cpuidle/state1/name:C1 /sys/devices/system/cpu/cpu0/cpuidle/state1/s2idle/time:0 /sys/devices/system/cpu/cpu0/cpuidle/state1/s2idle/usage:0 Thanks!
On Wed, 2020-08-26 at 16:16 +0300, Artem Bityutskiy wrote: > Just get a reasonably new turbostat (it is part of the kernel tree, you > can compile it yourself) and run it for few seconds (like 'turbostat > sleep 10'), get the output (will be a lot of it), and we can check what > is actually going on with regards to C-states. Oh, and if you could do that with and without your patch, we could even compare things. But try to do it at least with the default (acpi_idle) configuration. Artem.
On Wed, 26 Aug 2020 at 15:09, Artem Bityutskiy <dedekind1@gmail.com> wrote: > > On Wed, 2020-08-26 at 15:03 +0200, Guilhem Lettron wrote: > > On my laptop, a Dell XPS 13 7390 2-in-1 with i7-1065G7, ACPI only > > report "C1_ACPI", "C2_ACPI" and "C3_ACPI". > > Did you try to dig into the BIOS menus and check if you can enable > more/deeper C-states? Yes everything is enabled relative to C-states. I find, at least, 1 people with same problem: https://community.clearlinux.org/t/intel-idle-on-10th-ice-lake/3939/9 > > Artem. > Guilhem Lettron
On Wed, Aug 26, 2020 at 3:09 PM Artem Bityutskiy <dedekind1@gmail.com> wrote: > > On Wed, 2020-08-26 at 15:03 +0200, Guilhem Lettron wrote: > > On Wed, 26 Aug 2020 at 14:43, Rafael J. Wysocki <rafael@kernel.org> wrote: > > > On Wed, Aug 26, 2020 at 2:05 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > > > Use the same C-states as SKL > > > > > > Why is this change needed? > > > > On my laptop, a Dell XPS 13 7390 2-in-1 with i7-1065G7, ACPI only > > report "C1_ACPI", "C2_ACPI" and "C3_ACPI". > > Did you try to dig into the BIOS menus and check if you can enable > more/deeper C-states? I would be surprised if there were settings for that in the BIOS, as Windows only expects #1, C2 and C3 AFAICS. It might be possible to disable C1E autopromotion, but that option is not likely to be available on a laptop. Cheers!
On Wed, Aug 26, 2020 at 3:18 PM Artem Bityutskiy <dedekind1@gmail.com> wrote: > > On Wed, 2020-08-26 at 16:16 +0300, Artem Bityutskiy wrote: > > Just get a reasonably new turbostat (it is part of the kernel tree, you > > can compile it yourself) and run it for few seconds (like 'turbostat > > sleep 10'), get the output (will be a lot of it), and we can check what > > is actually going on with regards to C-states. > > Oh, and if you could do that with and without your patch, we could even > compare things. But try to do it at least with the default (acpi_idle) > configuration. IMO it is way easier to get the information from sysfs.
On Wed, 26 Aug 2020 at 15:18, Artem Bityutskiy <dedekind1@gmail.com> wrote: > > On Wed, 2020-08-26 at 16:16 +0300, Artem Bityutskiy wrote: > > Just get a reasonably new turbostat (it is part of the kernel tree, you > > can compile it yourself) and run it for few seconds (like 'turbostat > > sleep 10'), get the output (will be a lot of it), and we can check what > > is actually going on with regards to C-states. > > Oh, and if you could do that with and without your patch, we could even > compare things. But try to do it at least with the default (acpi_idle) > configuration. with my patch: turbostat version 20.03.20 - Len Brown <lenb@kernel.org> CPUID(0): GenuineIntel 0x1b CPUID levels; 0x80000008 xlevels; family:model:stepping 0x6:7e:5 (6:126:5) CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM HT TM CPUID(6): APERF, TURBO, DTS, PTM, HWP, No-HWPnotify, HWPwindow, HWPepp, HWPpkg, EPB cpu2: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST MWAIT PREFETCH TURBO) CPUID(7): SGX cpu2: MSR_IA32_FEATURE_CONTROL: 0x00020005 (Locked ) CPUID(0x15): eax_crystal: 2 ebx_tsc: 78 ecx_crystal_hz: 38400000 TSC: 1497 MHz (38400000 Hz * 78 / 2 / 1000000) CPUID(0x16): base_mhz: 1500 max_mhz: 3900 bus_mhz: 100 cpu2: MSR_MISC_PWR_MGMT: 0x00401c40 (ENable-EIST_Coordination DISable-EPB DISable-OOB) RAPL: 17476 sec. Joule Counter Range, at 15 Watts cpu2: MSR_PLATFORM_INFO: 0x4043cf1810f00 4 * 100.0 = 400.0 MHz max efficiency frequency 15 * 100.0 = 1500.0 MHz base frequency cpu2: MSR_IA32_POWER_CTL: 0x0024005d (C1E auto-promotion: DISabled) cpu2: MSR_TURBO_RATIO_LIMIT: 0x2323232323232627 35 * 100.0 = 3500.0 MHz max turbo 8 active cores 35 * 100.0 = 3500.0 MHz max turbo 7 active cores 35 * 100.0 = 3500.0 MHz max turbo 6 active cores 35 * 100.0 = 3500.0 MHz max turbo 5 active cores 35 * 100.0 = 3500.0 MHz max turbo 4 active cores 35 * 100.0 = 3500.0 MHz max turbo 3 active cores 38 * 100.0 = 3800.0 MHz max turbo 2 active cores 39 * 100.0 = 3900.0 MHz max turbo 1 active cores cpu2: MSR_CONFIG_TDP_NOMINAL: 0x0000000d (base_ratio=13) cpu2: MSR_CONFIG_TDP_LEVEL_1: 0x000a0060 (PKG_MIN_PWR_LVL1=0 PKG_MAX_PWR_LVL1=0 LVL1_RATIO=10 PKG_TDP_LVL1=96) cpu2: MSR_CONFIG_TDP_LEVEL_2: 0x000f00c8 (PKG_MIN_PWR_LVL2=0 PKG_MAX_PWR_LVL2=0 LVL2_RATIO=15 PKG_TDP_LVL2=200) cpu2: MSR_CONFIG_TDP_CONTROL: 0x00000000 ( lock=0) cpu2: MSR_TURBO_ACTIVATION_RATIO: 0x0000000c (MAX_NON_TURBO_RATIO=12 lock=0) cpu2: MSR_PKG_CST_CONFIG_CONTROL: 0x74008008 (UNdemote-C1, demote-C1, locked, pkg-cstate-limit=8 (unlimited)) current_driver: intel_idle current_governor: menu current_governor_ro: menu cpu2: POLL: CPUIDLE CORE POLL IDLE cpu2: C1: MWAIT 0x00 cpu2: C1E: MWAIT 0x01 cpu2: C6: MWAIT 0x20 cpu2: C7s: MWAIT 0x33 cpu2: C8: MWAIT 0x40 cpu2: C9: MWAIT 0x50 cpu2: C10: MWAIT 0x60 cpu2: cpufreq driver: intel_cpufreq cpu2: cpufreq governor: schedutil cpufreq intel_pstate no_turbo: 0 cpu2: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch) cpu0: MSR_PM_ENABLE: 0x00000001 (HWP) cpu0: MSR_HWP_CAPABILITIES: 0x010e0d27 (high 39 guar 13 eff 14 low 1) cpu0: MSR_HWP_REQUEST: 0x80002727 (min 39 max 39 des 0 epp 0x80 window 0x0 pkg 0x0) cpu0: MSR_HWP_REQUEST_PKG: 0x8000ff01 (min 1 max 255 des 0 epp 0x80 window 0x0) cpu0: MSR_HWP_STATUS: 0x00000000 (No-Guaranteed_Perf_Change, No-Excursion_Min) cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced) cpu0: MSR_RAPL_POWER_UNIT: 0x000a0e03 (0.125000 Watts, 0.000061 Joules, 0.000977 sec.) cpu0: MSR_PKG_POWER_INFO: 0x00000078 (15 W TDP, RAPL 0 - 0 W, 0.000000 sec.) cpu0: MSR_PKG_POWER_LIMIT: 0x5a8118009d80c8 (UNlocked) cpu0: PKG Limit #1: ENabled (25.000000 Watts, 24.000000 sec, clamp ENabled) cpu0: PKG Limit #2: ENabled (35.000000 Watts, 10.000000* sec, clamp DISabled) cpu0: MSR_DRAM_POWER_LIMIT: 0x5400de00000000 (UNlocked) cpu0: DRAM Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP0_POLICY: 0 cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked) cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP1_POLICY: 0 cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked) cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x05640000 (100 C) cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x88290800 (59 C) cpu0: MSR_IA32_PACKAGE_THERM_INTERRUPT: 0x00000003 (100 C, 100 C) cpu2: MSR_PKGC3_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC6_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC7_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC8_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC9_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC10_IRTL: 0x00000000 (NOTvalid, 0 ns) 10.003466 sec Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz IRQ SMI POLL C1 C1E C6 C7s C8 C9 C10 POLL% C1% C1E% C6% C7s% C8% C9% C10% CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 GFXMHz Totl%C0 Any%C0 GFX%C0 CPUGFX% Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 CPU%LPI SYS%LPI PkgWatt CorWatt GFXWatt RAMWatt PKG_% RAM_% - - 219 8.59 2549 1498 48489 160 71 1256 11389 21414 5 31317 489 8252 0.00 0.10 1.61 17.26 0.01 52.02 0.53 19.27 16.07 26.24 49.10 56 56 96.64 300 68.29 48.58 3.08 2.10 30.36 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.38 4.91 0.07 0.00 0.00 0.00 0 0 189 5.83 3239 1498 6131 20 26 222 1922 2648 0 4232 0 298 0.00 0.20 1.97 16.85 0.00 69.76 0.00 4.51 20.30 20.57 53.30 56 56 96.64 300 68.29 48.58 3.08 2.10 30.36 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.38 4.91 0.07 0.00 0.00 0.00 0 4 397 12.02 3307 1498 4248 20 5 173 1252 1552 0 3266 2 1367 0.00 0.08 1.22 10.13 0.00 46.47 0.01 28.39 14.11 1 1 302 8.53 3545 1498 6108 20 11 155 1215 3588 0 4462 0 2 0.00 0.11 1.90 28.91 0.00 59.04 0.00 0.08 14.13 31.04 46.31 55 1 5 149 6.29 2378 1498 3953 20 8 144 1391 1710 0 3239 9 1304 0.00 0.06 1.48 10.45 0.00 49.50 0.23 31.45 16.37 2 2 159 7.86 2022 1498 3688 20 5 111 1088 1450 0 3122 1 1191 0.00 0.14 0.95 9.49 0.00 49.73 0.07 31.64 13.61 19.24 59.29 54 2 6 175 7.44 2355 1498 6107 20 4 137 1726 2424 0 3863 4 651 0.00 0.06 1.90 15.83 0.00 56.39 0.04 18.00 14.03 3 3 213 11.84 1797 1498 13814 20 7 170 1431 6125 2 5634 464 2422 0.00 0.09 1.89 34.67 0.01 31.54 3.82 16.28 16.54 34.11 37.51 52 3 7 167 8.91 1871 1498 4440 20 5 144 1364 1917 3 3499 9 1017 0.00 0.07 1.55 11.78 0.03 53.73 0.07 23.83 19.47 without (3.9-rc2): turbostat version 20.03.20 - Len Brown <lenb@kernel.org> CPUID(0): GenuineIntel 0x1b CPUID levels; 0x80000008 xlevels; family:model:stepping 0x6:7e:5 (6:126:5) CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM HT TM CPUID(6): APERF, TURBO, DTS, PTM, HWP, No-HWPnotify, HWPwindow, HWPepp, HWPpkg, EPB cpu1: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST MWAIT PREFETCH TURBO) CPUID(7): SGX cpu1: MSR_IA32_FEATURE_CONTROL: 0x00020005 (Locked ) CPUID(0x15): eax_crystal: 2 ebx_tsc: 78 ecx_crystal_hz: 38400000 TSC: 1497 MHz (38400000 Hz * 78 / 2 / 1000000) CPUID(0x16): base_mhz: 1500 max_mhz: 3900 bus_mhz: 100 cpu1: MSR_MISC_PWR_MGMT: 0x00401c40 (ENable-EIST_Coordination DISable-EPB DISable-OOB) RAPL: 17476 sec. Joule Counter Range, at 15 Watts cpu1: MSR_PLATFORM_INFO: 0x4043cf1810f00 4 * 100.0 = 400.0 MHz max efficiency frequency 15 * 100.0 = 1500.0 MHz base frequency cpu1: MSR_IA32_POWER_CTL: 0x0024005f (C1E auto-promotion: ENabled) cpu1: MSR_TURBO_RATIO_LIMIT: 0x2323232323232627 35 * 100.0 = 3500.0 MHz max turbo 8 active cores 35 * 100.0 = 3500.0 MHz max turbo 7 active cores 35 * 100.0 = 3500.0 MHz max turbo 6 active cores 35 * 100.0 = 3500.0 MHz max turbo 5 active cores 35 * 100.0 = 3500.0 MHz max turbo 4 active cores 35 * 100.0 = 3500.0 MHz max turbo 3 active cores 38 * 100.0 = 3800.0 MHz max turbo 2 active cores 39 * 100.0 = 3900.0 MHz max turbo 1 active cores cpu1: MSR_CONFIG_TDP_NOMINAL: 0x0000000d (base_ratio=13) cpu1: MSR_CONFIG_TDP_LEVEL_1: 0x000a0060 (PKG_MIN_PWR_LVL1=0 PKG_MAX_PWR_LVL1=0 LVL1_RATIO=10 PKG_TDP_LVL1=96) cpu1: MSR_CONFIG_TDP_LEVEL_2: 0x000f00c8 (PKG_MIN_PWR_LVL2=0 PKG_MAX_PWR_LVL2=0 LVL2_RATIO=15 PKG_TDP_LVL2=200) cpu1: MSR_CONFIG_TDP_CONTROL: 0x00000000 ( lock=0) cpu1: MSR_TURBO_ACTIVATION_RATIO: 0x0000000c (MAX_NON_TURBO_RATIO=12 lock=0) cpu1: MSR_PKG_CST_CONFIG_CONTROL: 0x74008008 (UNdemote-C1, demote-C1, locked, pkg-cstate-limit=8 (unlimited)) current_driver: intel_idle current_governor: menu current_governor_ro: menu cpu1: POLL: CPUIDLE CORE POLL IDLE cpu1: C1_ACPI: ACPI FFH MWAIT 0x0 cpu1: C2_ACPI: ACPI FFH MWAIT 0x31 cpu1: C3_ACPI: ACPI FFH MWAIT 0x60 cpu1: cpufreq driver: intel_cpufreq cpu1: cpufreq governor: schedutil cpufreq intel_pstate no_turbo: 0 cpu1: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch) cpu0: MSR_PM_ENABLE: 0x00000001 (HWP) cpu0: MSR_HWP_CAPABILITIES: 0x010e0d27 (high 39 guar 13 eff 14 low 1) cpu0: MSR_HWP_REQUEST: 0x80002727 (min 39 max 39 des 0 epp 0x80 window 0x0 pkg 0x0) cpu0: MSR_HWP_REQUEST_PKG: 0x8000ff01 (min 1 max 255 des 0 epp 0x80 window 0x0) cpu0: MSR_HWP_STATUS: 0x00000000 (No-Guaranteed_Perf_Change, No-Excursion_Min) cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced) cpu0: MSR_RAPL_POWER_UNIT: 0x000a0e03 (0.125000 Watts, 0.000061 Joules, 0.000977 sec.) cpu0: MSR_PKG_POWER_INFO: 0x00000078 (15 W TDP, RAPL 0 - 0 W, 0.000000 sec.) cpu0: MSR_PKG_POWER_LIMIT: 0x42817000dd8170 (UNlocked) cpu0: PKG Limit #1: ENabled (46.000000 Watts, 28.000000 sec, clamp ENabled) cpu0: PKG Limit #2: ENabled (46.000000 Watts, 0.002441* sec, clamp DISabled) cpu0: MSR_DRAM_POWER_LIMIT: 0x5400de00000000 (UNlocked) cpu0: DRAM Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP0_POLICY: 0 cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked) cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP1_POLICY: 0 cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked) cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x00640000 (100 C) cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x881d0800 (71 C) cpu0: MSR_IA32_PACKAGE_THERM_INTERRUPT: 0x00000003 (100 C, 100 C) cpu1: MSR_PKGC3_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC6_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC7_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC8_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC9_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC10_IRTL: 0x00000000 (NOTvalid, 0 ns) 10.002424 sec Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz IRQ SMI POLL C1_ACPI C2_ACPI C3_ACPI POLL% C1_ACPI% C2_ACPI% C3_ACPI% CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 GFXMHz Totl%C0 Any%C0 GFX%C0 CPUGFX% Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 CPU%LPI SYS%LPI PkgWatt CorWatt GFXWatt RAMWatt PKG_% RAM_% - - 204 9.81 2083 1497 50389 176 820 25733 20628 14751 0.01 10.71 36.19 42.84 29.48 0.00 60.71 58 58 97.96 300 73.65 52.84 1.861.41 16.96 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.08 3.79 0.03 0.00 0.00 0.00 0 0 218 7.90 2760 1497 6824 22 4 3882 3146 864 0.00 12.68 53.86 24.76 26.32 0.00 65.77 58 58 97.96 300 73.65 52.84 1.861.41 16.96 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.08 3.79 0.03 0.00 0.00 0.00 0 4 144 6.91 2084 1497 4409 22 7 2410 2078 1608 0.00 7.55 32.26 52.88 27.32 1 1 154 8.09 1903 1497 4783 22 3 2528 2178 1801 0.00 7.54 34.66 49.51 42.41 0.00 49.50 57 1 5 290 14.49 2000 1497 14750 22 755 5964 4261 4558 0.04 22.47 27.57 35.05 36.01 2 2 209 11.35 1840 1497 4663 22 1 2771 2132 1467 0.00 8.39 31.37 48.69 28.70 0.00 59.96 55 2 6 232 13.35 1738 1497 5625 22 40 3313 2140 1338 0.00 10.36 33.13 42.87 26.70 3 3 221 9.01 2451 1497 4944 22 7 2627 2536 1447 0.00 9.32 41.41 39.55 23.38 0.00 67.61 54 3 7 167 7.39 2266 1497 4391 22 3 2238 2157 1668 0.00 7.39 35.25 49.42 25.00 > > Artem. > Guilhem Lettron
On Wed, 26 Aug 2020 at 15:17, Rafael J. Wysocki <rafael@kernel.org> wrote: > > On Wed, Aug 26, 2020 at 2:59 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > > > On my laptop, a Dell XPS 13 7390 2-in-1 with i7-1065G7, ACPI only report "C1_ACPI", "C2_ACPI" and "C3_ACPI". > > Can you list the properties of the idle states in this configuration, that is: > > $ grep -r '.*' /sys/devices/system/cpu/cpu0/cpuidle/state1/ > > and analogously for state2 and state3, and send the output of this? > > On my desktop system the above comment produces the following output: > > /sys/devices/system/cpu/cpu0/cpuidle/state1/disable:0 > /sys/devices/system/cpu/cpu0/cpuidle/state1/above:1022634 > /sys/devices/system/cpu/cpu0/cpuidle/state1/time:2556533185 > /sys/devices/system/cpu/cpu0/cpuidle/state1/power:0 > /sys/devices/system/cpu/cpu0/cpuidle/state1/residency:2 > /sys/devices/system/cpu/cpu0/cpuidle/state1/latency:2 > /sys/devices/system/cpu/cpu0/cpuidle/state1/usage:17625391 > /sys/devices/system/cpu/cpu0/cpuidle/state1/desc:MWAIT 0x00 > /sys/devices/system/cpu/cpu0/cpuidle/state1/below:6578613 > /sys/devices/system/cpu/cpu0/cpuidle/state1/default_status:enabled > /sys/devices/system/cpu/cpu0/cpuidle/state1/name:C1 > /sys/devices/system/cpu/cpu0/cpuidle/state1/s2idle/time:0 > /sys/devices/system/cpu/cpu0/cpuidle/state1/s2idle/usage:0 > > Thanks! grep -r '.*' /sys/devices/system/cpu/cpu0/cpuidle/state1/ /sys/devices/system/cpu/cpu0/cpuidle/state1/disable:0 /sys/devices/system/cpu/cpu0/cpuidle/state1/above:524 /sys/devices/system/cpu/cpu0/cpuidle/state1/time:63400881 /sys/devices/system/cpu/cpu0/cpuidle/state1/power:0 /sys/devices/system/cpu/cpu0/cpuidle/state1/residency:1 /sys/devices/system/cpu/cpu0/cpuidle/state1/latency:1 /sys/devices/system/cpu/cpu0/cpuidle/state1/usage:195432 /sys/devices/system/cpu/cpu0/cpuidle/state1/desc:ACPI FFH MWAIT 0x0 /sys/devices/system/cpu/cpu0/cpuidle/state1/below:27374 /sys/devices/system/cpu/cpu0/cpuidle/state1/default_status:enabled /sys/devices/system/cpu/cpu0/cpuidle/state1/name:C1_ACPI /sys/devices/system/cpu/cpu0/cpuidle/state1/s2idle/time:0 /sys/devices/system/cpu/cpu0/cpuidle/state1/s2idle/usage:0 grep -r '.*' /sys/devices/system/cpu/cpu0/cpuidle/state2/ /sys/devices/system/cpu/cpu0/cpuidle/state2/disable:0 /sys/devices/system/cpu/cpu0/cpuidle/state2/above:43616 /sys/devices/system/cpu/cpu0/cpuidle/state2/time:144893598 /sys/devices/system/cpu/cpu0/cpuidle/state2/power:0 /sys/devices/system/cpu/cpu0/cpuidle/state2/residency:759 /sys/devices/system/cpu/cpu0/cpuidle/state2/latency:253 /sys/devices/system/cpu/cpu0/cpuidle/state2/usage:102740 /sys/devices/system/cpu/cpu0/cpuidle/state2/desc:ACPI FFH MWAIT 0x31 /sys/devices/system/cpu/cpu0/cpuidle/state2/below:11351 /sys/devices/system/cpu/cpu0/cpuidle/state2/default_status:enabled /sys/devices/system/cpu/cpu0/cpuidle/state2/name:C2_ACPI /sys/devices/system/cpu/cpu0/cpuidle/state2/s2idle/time:0 /sys/devices/system/cpu/cpu0/cpuidle/state2/s2idle/usage:0 grep -r '.*' /sys/devices/system/cpu/cpu0/cpuidle/state3/ /sys/devices/system/cpu/cpu0/cpuidle/state3/disable:0 /sys/devices/system/cpu/cpu0/cpuidle/state3/above:18551 /sys/devices/system/cpu/cpu0/cpuidle/state3/time:43225299 /sys/devices/system/cpu/cpu0/cpuidle/state3/power:0 /sys/devices/system/cpu/cpu0/cpuidle/state3/residency:3144 /sys/devices/system/cpu/cpu0/cpuidle/state3/latency:1048 /sys/devices/system/cpu/cpu0/cpuidle/state3/usage:24205 /sys/devices/system/cpu/cpu0/cpuidle/state3/desc:ACPI FFH MWAIT 0x60 /sys/devices/system/cpu/cpu0/cpuidle/state3/below:0 /sys/devices/system/cpu/cpu0/cpuidle/state3/default_status:enabled /sys/devices/system/cpu/cpu0/cpuidle/state3/name:C3_ACPI /sys/devices/system/cpu/cpu0/cpuidle/state3/s2idle/time:0 /sys/devices/system/cpu/cpu0/cpuidle/state3/s2idle/usage:0 Guilhem Lettron
On Wed, 2020-08-26 at 15:32 +0200, Guilhem Lettron wrote: > On Wed, 26 Aug 2020 at 15:18, Artem Bityutskiy <dedekind1@gmail.com> > wrote: > > > > On Wed, 2020-08-26 at 16:16 +0300, Artem Bityutskiy wrote: > > > Just get a reasonably new turbostat (it is part of the kernel > > > tree, you > > > can compile it yourself) and run it for few seconds (like > > > 'turbostat > > > sleep 10'), get the output (will be a lot of it), and we can > > > check what > > > is actually going on with regards to C-states. > > > > Oh, and if you could do that with and without your patch, we could > > even > > compare things. But try to do it at least with the default > > (acpi_idle) > > configuration. > > with my patch: > > turbostat version 20.03.20 - Len Brown <lenb@kernel.org> > CPUID(0): GenuineIntel 0x1b CPUID levels; 0x80000008 xlevels; > family:model:stepping 0x6:7e:5 (6:126:5) > CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM HT TM > CPUID(6): APERF, TURBO, DTS, PTM, HWP, No-HWPnotify, HWPwindow, > HWPepp, HWPpkg, EPB > cpu2: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST MWAIT PREFETCH > TURBO) > CPUID(7): SGX > cpu2: MSR_IA32_FEATURE_CONTROL: 0x00020005 (Locked ) > CPUID(0x15): eax_crystal: 2 ebx_tsc: 78 ecx_crystal_hz: 38400000 > TSC: 1497 MHz (38400000 Hz * 78 / 2 / 1000000) > CPUID(0x16): base_mhz: 1500 max_mhz: 3900 bus_mhz: 100 > cpu2: MSR_MISC_PWR_MGMT: 0x00401c40 (ENable-EIST_Coordination > DISable-EPB DISable-OOB) > RAPL: 17476 sec. Joule Counter Range, at 15 Watts > cpu2: MSR_PLATFORM_INFO: 0x4043cf1810f00 > 4 * 100.0 = 400.0 MHz max efficiency frequency > 15 * 100.0 = 1500.0 MHz base frequency > cpu2: MSR_IA32_POWER_CTL: 0x0024005d (C1E auto-promotion: DISabled) > cpu2: MSR_TURBO_RATIO_LIMIT: 0x2323232323232627 > 35 * 100.0 = 3500.0 MHz max turbo 8 active cores > 35 * 100.0 = 3500.0 MHz max turbo 7 active cores > 35 * 100.0 = 3500.0 MHz max turbo 6 active cores > 35 * 100.0 = 3500.0 MHz max turbo 5 active cores > 35 * 100.0 = 3500.0 MHz max turbo 4 active cores > 35 * 100.0 = 3500.0 MHz max turbo 3 active cores > 38 * 100.0 = 3800.0 MHz max turbo 2 active cores > 39 * 100.0 = 3900.0 MHz max turbo 1 active cores > cpu2: MSR_CONFIG_TDP_NOMINAL: 0x0000000d (base_ratio=13) > cpu2: MSR_CONFIG_TDP_LEVEL_1: 0x000a0060 (PKG_MIN_PWR_LVL1=0 > PKG_MAX_PWR_LVL1=0 LVL1_RATIO=10 PKG_TDP_LVL1=96) > cpu2: MSR_CONFIG_TDP_LEVEL_2: 0x000f00c8 (PKG_MIN_PWR_LVL2=0 > PKG_MAX_PWR_LVL2=0 LVL2_RATIO=15 PKG_TDP_LVL2=200) > cpu2: MSR_CONFIG_TDP_CONTROL: 0x00000000 ( lock=0) > cpu2: MSR_TURBO_ACTIVATION_RATIO: 0x0000000c (MAX_NON_TURBO_RATIO=12 > lock=0) > cpu2: MSR_PKG_CST_CONFIG_CONTROL: 0x74008008 (UNdemote-C1, demote-C1, > locked, pkg-cstate-limit=8 (unlimited)) > current_driver: intel_idle > current_governor: menu > current_governor_ro: menu > cpu2: POLL: CPUIDLE CORE POLL IDLE > cpu2: C1: MWAIT 0x00 > cpu2: C1E: MWAIT 0x01 > cpu2: C6: MWAIT 0x20 > cpu2: C7s: MWAIT 0x33 > cpu2: C8: MWAIT 0x40 > cpu2: C9: MWAIT 0x50 > cpu2: C10: MWAIT 0x60 > cpu2: cpufreq driver: intel_cpufreq > cpu2: cpufreq governor: schedutil > cpufreq intel_pstate no_turbo: 0 > cpu2: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch > L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch) > cpu0: MSR_PM_ENABLE: 0x00000001 (HWP) > cpu0: MSR_HWP_CAPABILITIES: 0x010e0d27 (high 39 guar 13 eff 14 low 1) > cpu0: MSR_HWP_REQUEST: 0x80002727 (min 39 max 39 des 0 epp 0x80 > window > 0x0 pkg 0x0) > cpu0: MSR_HWP_REQUEST_PKG: 0x8000ff01 (min 1 max 255 des 0 epp 0x80 > window 0x0) > cpu0: MSR_HWP_STATUS: 0x00000000 (No-Guaranteed_Perf_Change, No- > Excursion_Min) > cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced) > cpu0: MSR_RAPL_POWER_UNIT: 0x000a0e03 (0.125000 Watts, 0.000061 > Joules, 0.000977 sec.) > cpu0: MSR_PKG_POWER_INFO: 0x00000078 (15 W TDP, RAPL 0 - 0 W, > 0.000000 sec.) > cpu0: MSR_PKG_POWER_LIMIT: 0x5a8118009d80c8 (UNlocked) > cpu0: PKG Limit #1: ENabled (25.000000 Watts, 24.000000 sec, clamp > ENabled) > cpu0: PKG Limit #2: ENabled (35.000000 Watts, 10.000000* sec, clamp > DISabled) > cpu0: MSR_DRAM_POWER_LIMIT: 0x5400de00000000 (UNlocked) > cpu0: DRAM Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp > DISabled) > cpu0: MSR_PP0_POLICY: 0 > cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked) > cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp > DISabled) > cpu0: MSR_PP1_POLICY: 0 > cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked) > cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp > DISabled) > cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x05640000 (100 C) > cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x88290800 (59 C) > cpu0: MSR_IA32_PACKAGE_THERM_INTERRUPT: 0x00000003 (100 C, 100 C) > cpu2: MSR_PKGC3_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu2: MSR_PKGC6_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu2: MSR_PKGC7_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu2: MSR_PKGC8_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu2: MSR_PKGC9_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu2: MSR_PKGC10_IRTL: 0x00000000 (NOTvalid, 0 ns) > 10.003466 sec > Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz IRQ SMI POLL C1 C1E C6 C7s C8 > C9 C10 POLL% C1% C1E% C6% C7s% C8% C9% C10% CPU%c1 CPU%c6 CPU%c7 > CoreTmp PkgTmp GFX%rc6 GFXMHz Totl%C0 Any%C0 GFX%C0 CPUGFX% Pkg%pc2 > Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 CPU%LPI SYS%LPI > PkgWatt CorWatt GFXWatt RAMWatt PKG_% RAM_% This is really hard to read. can you please attach the two turbostat output as attachments? thanks, rui > - - 219 8.59 2549 1498 48489 160 71 1256 11389 21414 5 31317 489 8252 > 0.00 0.10 1.61 17.26 0.01 52.02 0.53 19.27 16.07 26.24 49.10 56 56 > 96.64 300 68.29 48.58 3.08 2.10 30.36 0.04 0.00 0.00 0.00 0.00 0.00 > 0.00 0.00 7.38 4.91 0.07 0.00 0.00 0.00 > 0 0 189 5.83 3239 1498 6131 20 26 222 1922 2648 0 4232 0 298 0.00 > 0.20 > 1.97 16.85 0.00 69.76 0.00 4.51 20.30 20.57 53.30 56 56 96.64 300 > 68.29 48.58 3.08 2.10 30.36 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 > 7.38 4.91 0.07 0.00 0.00 0.00 > 0 4 397 12.02 3307 1498 4248 20 5 173 1252 1552 0 3266 2 1367 0.00 > 0.08 1.22 10.13 0.00 46.47 0.01 28.39 14.11 > 1 1 302 8.53 3545 1498 6108 20 11 155 1215 3588 0 4462 0 2 0.00 0.11 > 1.90 28.91 0.00 59.04 0.00 0.08 14.13 31.04 46.31 55 > 1 5 149 6.29 2378 1498 3953 20 8 144 1391 1710 0 3239 9 1304 0.00 > 0.06 > 1.48 10.45 0.00 49.50 0.23 31.45 16.37 > 2 2 159 7.86 2022 1498 3688 20 5 111 1088 1450 0 3122 1 1191 0.00 > 0.14 > 0.95 9.49 0.00 49.73 0.07 31.64 13.61 19.24 59.29 54 > 2 6 175 7.44 2355 1498 6107 20 4 137 1726 2424 0 3863 4 651 0.00 0.06 > 1.90 15.83 0.00 56.39 0.04 18.00 14.03 > 3 3 213 11.84 1797 1498 13814 20 7 170 1431 6125 2 5634 464 2422 0.00 > 0.09 1.89 34.67 0.01 31.54 3.82 16.28 16.54 34.11 37.51 52 > 3 7 167 8.91 1871 1498 4440 20 5 144 1364 1917 3 3499 9 1017 0.00 > 0.07 > 1.55 11.78 0.03 53.73 0.07 23.83 19.47 > > without (3.9-rc2): > > turbostat version 20.03.20 - Len Brown <lenb@kernel.org> > CPUID(0): GenuineIntel 0x1b CPUID levels; 0x80000008 xlevels; > family:model:stepping 0x6:7e:5 (6:126:5) > CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM HT TM > CPUID(6): APERF, TURBO, DTS, PTM, HWP, No-HWPnotify, HWPwindow, > HWPepp, HWPpkg, EPB > cpu1: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST MWAIT PREFETCH > TURBO) > CPUID(7): SGX > cpu1: MSR_IA32_FEATURE_CONTROL: 0x00020005 (Locked ) > CPUID(0x15): eax_crystal: 2 ebx_tsc: 78 ecx_crystal_hz: 38400000 > TSC: 1497 MHz (38400000 Hz * 78 / 2 / 1000000) > CPUID(0x16): base_mhz: 1500 max_mhz: 3900 bus_mhz: 100 > cpu1: MSR_MISC_PWR_MGMT: 0x00401c40 (ENable-EIST_Coordination > DISable-EPB DISable-OOB) > RAPL: 17476 sec. Joule Counter Range, at 15 Watts > cpu1: MSR_PLATFORM_INFO: 0x4043cf1810f00 > 4 * 100.0 = 400.0 MHz max efficiency frequency > 15 * 100.0 = 1500.0 MHz base frequency > cpu1: MSR_IA32_POWER_CTL: 0x0024005f (C1E auto-promotion: ENabled) > cpu1: MSR_TURBO_RATIO_LIMIT: 0x2323232323232627 > 35 * 100.0 = 3500.0 MHz max turbo 8 active cores > 35 * 100.0 = 3500.0 MHz max turbo 7 active cores > 35 * 100.0 = 3500.0 MHz max turbo 6 active cores > 35 * 100.0 = 3500.0 MHz max turbo 5 active cores > 35 * 100.0 = 3500.0 MHz max turbo 4 active cores > 35 * 100.0 = 3500.0 MHz max turbo 3 active cores > 38 * 100.0 = 3800.0 MHz max turbo 2 active cores > 39 * 100.0 = 3900.0 MHz max turbo 1 active cores > cpu1: MSR_CONFIG_TDP_NOMINAL: 0x0000000d (base_ratio=13) > cpu1: MSR_CONFIG_TDP_LEVEL_1: 0x000a0060 (PKG_MIN_PWR_LVL1=0 > PKG_MAX_PWR_LVL1=0 LVL1_RATIO=10 PKG_TDP_LVL1=96) > cpu1: MSR_CONFIG_TDP_LEVEL_2: 0x000f00c8 (PKG_MIN_PWR_LVL2=0 > PKG_MAX_PWR_LVL2=0 LVL2_RATIO=15 PKG_TDP_LVL2=200) > cpu1: MSR_CONFIG_TDP_CONTROL: 0x00000000 ( lock=0) > cpu1: MSR_TURBO_ACTIVATION_RATIO: 0x0000000c (MAX_NON_TURBO_RATIO=12 > lock=0) > cpu1: MSR_PKG_CST_CONFIG_CONTROL: 0x74008008 (UNdemote-C1, demote-C1, > locked, pkg-cstate-limit=8 (unlimited)) > current_driver: intel_idle > current_governor: menu > current_governor_ro: menu > cpu1: POLL: CPUIDLE CORE POLL IDLE > cpu1: C1_ACPI: ACPI FFH MWAIT 0x0 > cpu1: C2_ACPI: ACPI FFH MWAIT 0x31 > cpu1: C3_ACPI: ACPI FFH MWAIT 0x60 > cpu1: cpufreq driver: intel_cpufreq > cpu1: cpufreq governor: schedutil > cpufreq intel_pstate no_turbo: 0 > cpu1: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch > L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch) > cpu0: MSR_PM_ENABLE: 0x00000001 (HWP) > cpu0: MSR_HWP_CAPABILITIES: 0x010e0d27 (high 39 guar 13 eff 14 low 1) > cpu0: MSR_HWP_REQUEST: 0x80002727 (min 39 max 39 des 0 epp 0x80 > window > 0x0 pkg 0x0) > cpu0: MSR_HWP_REQUEST_PKG: 0x8000ff01 (min 1 max 255 des 0 epp 0x80 > window 0x0) > cpu0: MSR_HWP_STATUS: 0x00000000 (No-Guaranteed_Perf_Change, No- > Excursion_Min) > cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced) > cpu0: MSR_RAPL_POWER_UNIT: 0x000a0e03 (0.125000 Watts, 0.000061 > Joules, 0.000977 sec.) > cpu0: MSR_PKG_POWER_INFO: 0x00000078 (15 W TDP, RAPL 0 - 0 W, > 0.000000 sec.) > cpu0: MSR_PKG_POWER_LIMIT: 0x42817000dd8170 (UNlocked) > cpu0: PKG Limit #1: ENabled (46.000000 Watts, 28.000000 sec, clamp > ENabled) > cpu0: PKG Limit #2: ENabled (46.000000 Watts, 0.002441* sec, clamp > DISabled) > cpu0: MSR_DRAM_POWER_LIMIT: 0x5400de00000000 (UNlocked) > cpu0: DRAM Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp > DISabled) > cpu0: MSR_PP0_POLICY: 0 > cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked) > cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp > DISabled) > cpu0: MSR_PP1_POLICY: 0 > cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked) > cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp > DISabled) > cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x00640000 (100 C) > cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x881d0800 (71 C) > cpu0: MSR_IA32_PACKAGE_THERM_INTERRUPT: 0x00000003 (100 C, 100 C) > cpu1: MSR_PKGC3_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu1: MSR_PKGC6_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu1: MSR_PKGC7_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu1: MSR_PKGC8_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu1: MSR_PKGC9_IRTL: 0x00000000 (NOTvalid, 0 ns) > cpu1: MSR_PKGC10_IRTL: 0x00000000 (NOTvalid, 0 ns) > 10.002424 sec > Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz IRQ SMI POLL C1_ACPI C2_ACPI > C3_ACPI POLL% C1_ACPI% C2_ACPI% C3_ACPI% CPU%c1 CPU%c6 CPU%c7 CoreTmp > PkgTmp GFX%rc6 GFXMHz Totl%C0 Any%C0 GFX%C0 CPUGFX% Pkg%pc2 Pkg%pc3 > Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 CPU%LPI SYS%LPI PkgWatt > CorWatt GFXWatt RAMWatt PKG_% RAM_% > - - 204 9.81 2083 1497 50389 176 820 25733 20628 14751 0.01 10.71 > 36.19 42.84 29.48 0.00 60.71 58 58 97.96 300 73.65 52.84 1.861.41 > 16.96 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.08 3.79 0.03 0.00 > 0.00 > 0.00 > 0 0 218 7.90 2760 1497 6824 22 4 3882 3146 864 0.00 12.68 53.86 24.76 > 26.32 0.00 65.77 58 58 97.96 300 73.65 52.84 1.861.41 16.96 0.00 0.00 > 0.00 0.00 0.00 0.00 0.00 0.00 6.08 3.79 0.03 0.00 0.00 0.00 > 0 4 144 6.91 2084 1497 4409 22 7 2410 2078 1608 0.00 7.55 32.26 52.88 > 27.32 > 1 1 154 8.09 1903 1497 4783 22 3 2528 2178 1801 0.00 7.54 34.66 49.51 > 42.41 0.00 49.50 57 > 1 5 290 14.49 2000 1497 14750 22 755 5964 4261 4558 0.04 22.47 27.57 > 35.05 36.01 > 2 2 209 11.35 1840 1497 4663 22 1 2771 2132 1467 0.00 8.39 31.37 > 48.69 > 28.70 0.00 59.96 55 > 2 6 232 13.35 1738 1497 5625 22 40 3313 2140 1338 0.00 10.36 33.13 > 42.87 26.70 > 3 3 221 9.01 2451 1497 4944 22 7 2627 2536 1447 0.00 9.32 41.41 39.55 > 23.38 0.00 67.61 54 > 3 7 167 7.39 2266 1497 4391 22 3 2238 2157 1668 0.00 7.39 35.25 49.42 > 25.00 > > > > > > Artem. > > > > Guilhem Lettron
On Wed, 26 Aug 2020 at 15:41, Zhang Rui <rui.zhang@intel.com> wrote: > > > This is really hard to read. > can you please attach the two turbostat output as attachments? of course :) Guilhem Lettron turbostat version 20.03.20 - Len Brown <lenb@kernel.org> CPUID(0): GenuineIntel 0x1b CPUID levels; 0x80000008 xlevels; family:model:stepping 0x6:7e:5 (6:126:5) CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM HT TM CPUID(6): APERF, TURBO, DTS, PTM, HWP, No-HWPnotify, HWPwindow, HWPepp, HWPpkg, EPB cpu2: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST MWAIT PREFETCH TURBO) CPUID(7): SGX cpu2: MSR_IA32_FEATURE_CONTROL: 0x00020005 (Locked ) CPUID(0x15): eax_crystal: 2 ebx_tsc: 78 ecx_crystal_hz: 38400000 TSC: 1497 MHz (38400000 Hz * 78 / 2 / 1000000) CPUID(0x16): base_mhz: 1500 max_mhz: 3900 bus_mhz: 100 cpu2: MSR_MISC_PWR_MGMT: 0x00401c40 (ENable-EIST_Coordination DISable-EPB DISable-OOB) RAPL: 17476 sec. Joule Counter Range, at 15 Watts cpu2: MSR_PLATFORM_INFO: 0x4043cf1810f00 4 * 100.0 = 400.0 MHz max efficiency frequency 15 * 100.0 = 1500.0 MHz base frequency cpu2: MSR_IA32_POWER_CTL: 0x0024005d (C1E auto-promotion: DISabled) cpu2: MSR_TURBO_RATIO_LIMIT: 0x2323232323232627 35 * 100.0 = 3500.0 MHz max turbo 8 active cores 35 * 100.0 = 3500.0 MHz max turbo 7 active cores 35 * 100.0 = 3500.0 MHz max turbo 6 active cores 35 * 100.0 = 3500.0 MHz max turbo 5 active cores 35 * 100.0 = 3500.0 MHz max turbo 4 active cores 35 * 100.0 = 3500.0 MHz max turbo 3 active cores 38 * 100.0 = 3800.0 MHz max turbo 2 active cores 39 * 100.0 = 3900.0 MHz max turbo 1 active cores cpu2: MSR_CONFIG_TDP_NOMINAL: 0x0000000d (base_ratio=13) cpu2: MSR_CONFIG_TDP_LEVEL_1: 0x000a0060 (PKG_MIN_PWR_LVL1=0 PKG_MAX_PWR_LVL1=0 LVL1_RATIO=10 PKG_TDP_LVL1=96) cpu2: MSR_CONFIG_TDP_LEVEL_2: 0x000f00c8 (PKG_MIN_PWR_LVL2=0 PKG_MAX_PWR_LVL2=0 LVL2_RATIO=15 PKG_TDP_LVL2=200) cpu2: MSR_CONFIG_TDP_CONTROL: 0x00000000 ( lock=0) cpu2: MSR_TURBO_ACTIVATION_RATIO: 0x0000000c (MAX_NON_TURBO_RATIO=12 lock=0) cpu2: MSR_PKG_CST_CONFIG_CONTROL: 0x74008008 (UNdemote-C1, demote-C1, locked, pkg-cstate-limit=8 (unlimited)) current_driver: intel_idle current_governor: menu current_governor_ro: menu cpu2: POLL: CPUIDLE CORE POLL IDLE cpu2: C1: MWAIT 0x00 cpu2: C1E: MWAIT 0x01 cpu2: C6: MWAIT 0x20 cpu2: C7s: MWAIT 0x33 cpu2: C8: MWAIT 0x40 cpu2: C9: MWAIT 0x50 cpu2: C10: MWAIT 0x60 cpu2: cpufreq driver: intel_cpufreq cpu2: cpufreq governor: schedutil cpufreq intel_pstate no_turbo: 0 cpu2: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch) cpu0: MSR_PM_ENABLE: 0x00000001 (HWP) cpu0: MSR_HWP_CAPABILITIES: 0x010e0d27 (high 39 guar 13 eff 14 low 1) cpu0: MSR_HWP_REQUEST: 0x80002727 (min 39 max 39 des 0 epp 0x80 window 0x0 pkg 0x0) cpu0: MSR_HWP_REQUEST_PKG: 0x8000ff01 (min 1 max 255 des 0 epp 0x80 window 0x0) cpu0: MSR_HWP_STATUS: 0x00000000 (No-Guaranteed_Perf_Change, No-Excursion_Min) cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced) cpu0: MSR_RAPL_POWER_UNIT: 0x000a0e03 (0.125000 Watts, 0.000061 Joules, 0.000977 sec.) cpu0: MSR_PKG_POWER_INFO: 0x00000078 (15 W TDP, RAPL 0 - 0 W, 0.000000 sec.) cpu0: MSR_PKG_POWER_LIMIT: 0x5a8118009d80c8 (UNlocked) cpu0: PKG Limit #1: ENabled (25.000000 Watts, 24.000000 sec, clamp ENabled) cpu0: PKG Limit #2: ENabled (35.000000 Watts, 10.000000* sec, clamp DISabled) cpu0: MSR_DRAM_POWER_LIMIT: 0x5400de00000000 (UNlocked) cpu0: DRAM Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP0_POLICY: 0 cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked) cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP1_POLICY: 0 cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked) cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x05640000 (100 C) cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x88290800 (59 C) cpu0: MSR_IA32_PACKAGE_THERM_INTERRUPT: 0x00000003 (100 C, 100 C) cpu2: MSR_PKGC3_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC6_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC7_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC8_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC9_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu2: MSR_PKGC10_IRTL: 0x00000000 (NOTvalid, 0 ns) 10.003466 sec Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz IRQ SMI POLL C1 C1E C6 C7s C8 C9 C10 POLL% C1% C1E% C6% C7s% C8% C9% C10% CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 GFXMHz Totl%C0 Any%C0 GFX%C0 CPUGFX% Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 CPU%LPI SYS%LPI PkgWatt CorWatt GFXWatt RAMWatt PKG_% RAM_% - - 219 8.59 2549 1498 48489 160 71 1256 11389 21414 5 31317 489 8252 0.00 0.10 1.61 17.26 0.01 52.02 0.53 19.27 16.07 26.24 49.10 56 56 96.64 300 68.29 48.58 3.08 2.10 30.36 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.38 4.91 0.07 0.00 0.00 0.00 0 0 189 5.83 3239 1498 6131 20 26 222 1922 2648 0 4232 0 298 0.00 0.20 1.97 16.85 0.00 69.76 0.00 4.51 20.30 20.57 53.30 56 56 96.64 300 68.29 48.58 3.08 2.10 30.36 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.38 4.91 0.07 0.00 0.00 0.00 0 4 397 12.02 3307 1498 4248 20 5 173 1252 1552 0 3266 2 1367 0.00 0.08 1.22 10.13 0.00 46.47 0.01 28.39 14.11 1 1 302 8.53 3545 1498 6108 20 11 155 1215 3588 0 4462 0 2 0.00 0.11 1.90 28.91 0.00 59.04 0.00 0.08 14.13 31.04 46.31 55 1 5 149 6.29 2378 1498 3953 20 8 144 1391 1710 0 3239 9 1304 0.00 0.06 1.48 10.45 0.00 49.50 0.23 31.45 16.37 2 2 159 7.86 2022 1498 3688 20 5 111 1088 1450 0 3122 1 1191 0.00 0.14 0.95 9.49 0.00 49.73 0.07 31.64 13.61 19.24 59.29 54 2 6 175 7.44 2355 1498 6107 20 4 137 1726 2424 0 3863 4 651 0.00 0.06 1.90 15.83 0.00 56.39 0.04 18.00 14.03 3 3 213 11.84 1797 1498 13814 20 7 170 1431 6125 2 5634 464 2422 0.00 0.09 1.89 34.67 0.01 31.54 3.82 16.28 16.54 34.11 37.51 52 3 7 167 8.91 1871 1498 4440 20 5 144 1364 1917 3 3499 9 1017 0.00 0.07 1.55 11.78 0.03 53.73 0.07 23.83 19.47 turbostat version 20.03.20 - Len Brown <lenb@kernel.org> CPUID(0): GenuineIntel 0x1b CPUID levels; 0x80000008 xlevels; family:model:stepping 0x6:7e:5 (6:126:5) CPUID(1): SSE3 MONITOR - EIST TM2 TSC MSR ACPI-TM HT TM CPUID(6): APERF, TURBO, DTS, PTM, HWP, No-HWPnotify, HWPwindow, HWPepp, HWPpkg, EPB cpu1: MSR_IA32_MISC_ENABLE: 0x00850089 (TCC EIST MWAIT PREFETCH TURBO) CPUID(7): SGX cpu1: MSR_IA32_FEATURE_CONTROL: 0x00020005 (Locked ) CPUID(0x15): eax_crystal: 2 ebx_tsc: 78 ecx_crystal_hz: 38400000 TSC: 1497 MHz (38400000 Hz * 78 / 2 / 1000000) CPUID(0x16): base_mhz: 1500 max_mhz: 3900 bus_mhz: 100 cpu1: MSR_MISC_PWR_MGMT: 0x00401c40 (ENable-EIST_Coordination DISable-EPB DISable-OOB) RAPL: 17476 sec. Joule Counter Range, at 15 Watts cpu1: MSR_PLATFORM_INFO: 0x4043cf1810f00 4 * 100.0 = 400.0 MHz max efficiency frequency 15 * 100.0 = 1500.0 MHz base frequency cpu1: MSR_IA32_POWER_CTL: 0x0024005f (C1E auto-promotion: ENabled) cpu1: MSR_TURBO_RATIO_LIMIT: 0x2323232323232627 35 * 100.0 = 3500.0 MHz max turbo 8 active cores 35 * 100.0 = 3500.0 MHz max turbo 7 active cores 35 * 100.0 = 3500.0 MHz max turbo 6 active cores 35 * 100.0 = 3500.0 MHz max turbo 5 active cores 35 * 100.0 = 3500.0 MHz max turbo 4 active cores 35 * 100.0 = 3500.0 MHz max turbo 3 active cores 38 * 100.0 = 3800.0 MHz max turbo 2 active cores 39 * 100.0 = 3900.0 MHz max turbo 1 active cores cpu1: MSR_CONFIG_TDP_NOMINAL: 0x0000000d (base_ratio=13) cpu1: MSR_CONFIG_TDP_LEVEL_1: 0x000a0060 (PKG_MIN_PWR_LVL1=0 PKG_MAX_PWR_LVL1=0 LVL1_RATIO=10 PKG_TDP_LVL1=96) cpu1: MSR_CONFIG_TDP_LEVEL_2: 0x000f00c8 (PKG_MIN_PWR_LVL2=0 PKG_MAX_PWR_LVL2=0 LVL2_RATIO=15 PKG_TDP_LVL2=200) cpu1: MSR_CONFIG_TDP_CONTROL: 0x00000000 ( lock=0) cpu1: MSR_TURBO_ACTIVATION_RATIO: 0x0000000c (MAX_NON_TURBO_RATIO=12 lock=0) cpu1: MSR_PKG_CST_CONFIG_CONTROL: 0x74008008 (UNdemote-C1, demote-C1, locked, pkg-cstate-limit=8 (unlimited)) current_driver: intel_idle current_governor: menu current_governor_ro: menu cpu1: POLL: CPUIDLE CORE POLL IDLE cpu1: C1_ACPI: ACPI FFH MWAIT 0x0 cpu1: C2_ACPI: ACPI FFH MWAIT 0x31 cpu1: C3_ACPI: ACPI FFH MWAIT 0x60 cpu1: cpufreq driver: intel_cpufreq cpu1: cpufreq governor: schedutil cpufreq intel_pstate no_turbo: 0 cpu1: MSR_MISC_FEATURE_CONTROL: 0x00000000 (L2-Prefetch L2-Prefetch-pair L1-Prefetch L1-IP-Prefetch) cpu0: MSR_PM_ENABLE: 0x00000001 (HWP) cpu0: MSR_HWP_CAPABILITIES: 0x010e0d27 (high 39 guar 13 eff 14 low 1) cpu0: MSR_HWP_REQUEST: 0x80002727 (min 39 max 39 des 0 epp 0x80 window 0x0 pkg 0x0) cpu0: MSR_HWP_REQUEST_PKG: 0x8000ff01 (min 1 max 255 des 0 epp 0x80 window 0x0) cpu0: MSR_HWP_STATUS: 0x00000000 (No-Guaranteed_Perf_Change, No-Excursion_Min) cpu0: MSR_IA32_ENERGY_PERF_BIAS: 0x00000006 (balanced) cpu0: MSR_RAPL_POWER_UNIT: 0x000a0e03 (0.125000 Watts, 0.000061 Joules, 0.000977 sec.) cpu0: MSR_PKG_POWER_INFO: 0x00000078 (15 W TDP, RAPL 0 - 0 W, 0.000000 sec.) cpu0: MSR_PKG_POWER_LIMIT: 0x42817000dd8170 (UNlocked) cpu0: PKG Limit #1: ENabled (46.000000 Watts, 28.000000 sec, clamp ENabled) cpu0: PKG Limit #2: ENabled (46.000000 Watts, 0.002441* sec, clamp DISabled) cpu0: MSR_DRAM_POWER_LIMIT: 0x5400de00000000 (UNlocked) cpu0: DRAM Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP0_POLICY: 0 cpu0: MSR_PP0_POWER_LIMIT: 0x00000000 (UNlocked) cpu0: Cores Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_PP1_POLICY: 0 cpu0: MSR_PP1_POWER_LIMIT: 0x00000000 (UNlocked) cpu0: GFX Limit: DISabled (0.000000 Watts, 0.000977 sec, clamp DISabled) cpu0: MSR_IA32_TEMPERATURE_TARGET: 0x00640000 (100 C) cpu0: MSR_IA32_PACKAGE_THERM_STATUS: 0x881d0800 (71 C) cpu0: MSR_IA32_PACKAGE_THERM_INTERRUPT: 0x00000003 (100 C, 100 C) cpu1: MSR_PKGC3_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC6_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC7_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC8_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC9_IRTL: 0x00000000 (NOTvalid, 0 ns) cpu1: MSR_PKGC10_IRTL: 0x00000000 (NOTvalid, 0 ns) 10.002424 sec Core CPU Avg_MHz Busy% Bzy_MHz TSC_MHz IRQ SMI POLL C1_ACPI C2_ACPI C3_ACPI POLL% C1_ACPI% C2_ACPI% C3_ACPI% CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 GFXMHz Totl%C0 Any%C0 GFX%C0 CPUGFX% Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 CPU%LPI SYS%LPI PkgWatt CorWatt GFXWatt RAMWatt PKG_% RAM_% - - 204 9.81 2083 1497 50389 176 820 25733 20628 14751 0.01 10.71 36.19 42.84 29.48 0.00 60.71 58 58 97.96 300 73.65 52.84 1.861.41 16.96 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.08 3.79 0.03 0.00 0.00 0.00 0 0 218 7.90 2760 1497 6824 22 4 3882 3146 864 0.00 12.68 53.86 24.76 26.32 0.00 65.77 58 58 97.96 300 73.65 52.84 1.861.41 16.96 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.08 3.79 0.03 0.00 0.00 0.00 0 4 144 6.91 2084 1497 4409 22 7 2410 2078 1608 0.00 7.55 32.26 52.88 27.32 1 1 154 8.09 1903 1497 4783 22 3 2528 2178 1801 0.00 7.54 34.66 49.51 42.41 0.00 49.50 57 1 5 290 14.49 2000 1497 14750 22 755 5964 4261 4558 0.04 22.47 27.57 35.05 36.01 2 2 209 11.35 1840 1497 4663 22 1 2771 2132 1467 0.00 8.39 31.37 48.69 28.70 0.00 59.96 55 2 6 232 13.35 1738 1497 5625 22 40 3313 2140 1338 0.00 10.36 33.13 42.87 26.70 3 3 221 9.01 2451 1497 4944 22 7 2627 2536 1447 0.00 9.32 41.41 39.55 23.38 0.00 67.61 54 3 7 167 7.39 2266 1497 4391 22 3 2238 2157 1668 0.00 7.39 35.25 49.42 25.00
On Wed, Aug 26, 2020 at 4:04 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > On Wed, 26 Aug 2020 at 15:41, Zhang Rui <rui.zhang@intel.com> wrote: > > > > > > This is really hard to read. > > can you please attach the two turbostat output as attachments? > > of course :) Thanks! A couple of things happen here AFAICS. First, your processor seems to be unable to enter package C-states below PC3, so probably there is a device (most likely a PCI one) preventing it from doing that in the system. If all goes well, it should be able to get to at least PC8 without suspending the whole system. That needs to be dealt with in the first place before we can draw meaningful conclusions regarding which set of C-states to expose and whether or not the one exposed via ACPI is sufficient. To that end, I would try to upgrade the graphics firmware and see if you can get some nonzero PC8 residency then. Second, ACPI exposes C1, C7s and C10 only and so you don't get any CPU-C6 residency without the patch, but instead you get more CPU-C7 residency and more CPU-C1 residency. It is hard to say which is better in principle, but if you look at what is asked for by the governor, it turns out that deep C-states (C8-C10) are requested around 54% of the time with the patch, whereas without it the ACPI_C3 state (corresponding to C10) is requested approximately 24% of the time, which is much less often. That appears to translate to the difference in PC2 residency (~30% with the patch vs ~17% without it). Note, however, that (with the patch) C10 itself is asked for around 11% of the time which in turn is much less than the ~24% for the corresponding ACPI_C3 (without the patch). Overall, it looks like exposing C8 is beneficial from the energy usage perspective, because (in the future, when the "blocking" device is taken care of and the system can enter PC8 and deeper package C-states) it may allow PC8 to be entered more often in principle, even though it may reduce the amount of time spent in PC10 too (PC10 may be generally difficult to enter, though). [Here I'm assuming that the processor enters PC3 or PC2 instead of PC8 or deeper which cannot be entered due to some resource dependency.] OTOH exposing C1E doesn't seem to make much of a difference and exposing C6 only causes it to be asked for instead of C7s, so exposing the latter alone should be sufficient in theory. So IMO the set of C-states exposed by ACPI looks almost enough, but the jury is out until you can make the system be able to enter at least PC8. Cheers!
Indeed, when I compare them: acpi_idle (without the patch): CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 PkgWatt 29.48 0.00 60.71 58 58 97.96 16.96 0.00 0.00 0.00 0.00 0.00 0.00 6.08 intel_idle (with the patch): CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 PkgWatt 56 56 96.64 300 68.29 48.58 0.00 0.00 0.00 0.00 0.00 0.00 7.38 0.00 With intel_idle we reach PC10, without it we only go as deep as PC2 - huge difference. I really wonder why the BIOS does not expose deeper C-states... And if it does not, is this for a reason? And how windows works then? May be there is a BIOS update that fixes this problem? May be Windows user get it quickly because stuff like this is often well-integrated in Windows? Would you please check if there is newer BIOS? Artem.
On Wed, Aug 26, 2020 at 6:02 PM Rafael J. Wysocki <rafael@kernel.org> wrote: > > On Wed, Aug 26, 2020 at 4:04 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > > > On Wed, 26 Aug 2020 at 15:41, Zhang Rui <rui.zhang@intel.com> wrote: > > > > > > > > > This is really hard to read. > > > can you please attach the two turbostat output as attachments? > > > > of course :) > > Thanks! > > A couple of things happen here AFAICS. > > First, your processor seems to be unable to enter package C-states > below PC3, so probably there is a device (most likely a PCI one) > preventing it from doing that in the system. If all goes well, it > should be able to get to at least PC8 without suspending the whole > system. That needs to be dealt with in the first place before we can > draw meaningful conclusions regarding which set of C-states to expose > and whether or not the one exposed via ACPI is sufficient. > > To that end, I would try to upgrade the graphics firmware and see if > you can get some nonzero PC8 residency then. > > Second, ACPI exposes C1, C7s and C10 only and so you don't get any > CPU-C6 residency without the patch, but instead you get more CPU-C7 > residency and more CPU-C1 residency. It is hard to say which is > better in principle, but if you look at what is asked for by the > governor, it turns out that deep C-states (C8-C10) are requested > around 54% of the time with the patch, whereas without it the ACPI_C3 > state (corresponding to C10) is requested approximately 24% of the > time, which is much less often. That appears to translate to the > difference in PC2 residency (~30% with the patch vs ~17% without it). > > Note, however, that (with the patch) C10 itself is asked for around > 11% of the time which in turn is much less than the ~24% for the > corresponding ACPI_C3 (without the patch). > > Overall, it looks like exposing C8 is beneficial from the energy usage > perspective, because (in the future, when the "blocking" device is > taken care of and the system can enter PC8 and deeper package > C-states) it may allow PC8 to be entered more often in principle, even > though it may reduce the amount of time spent in PC10 too (PC10 may be > generally difficult to enter, though). [Here I'm assuming that the > processor enters PC3 or PC2 instead of PC8 or deeper which cannot be > entered due to some resource dependency.] > > OTOH exposing C1E doesn't seem to make much of a difference and > exposing C6 only causes it to be asked for instead of C7s, so exposing > the latter alone should be sufficient in theory. I need to correct myself here. With the patch C7s is exposed with the target residency of C8 which is why it is almost never selected by the governor and that's why C6 is kind of "instead" of it. So it looks like it might be beneficial to expose C8 (instead of C7s) and C6 with the target residency significantly less than that of C8. > So IMO the set of C-states exposed by ACPI looks almost enough, but > the jury is out until you can make the system be able to enter at > least PC8.
On Wed, 2020-08-26 at 19:19 +0300, Artem Bityutskiy wrote: > May be there is a BIOS update that fixes this problem? May be Windows > user get it quickly because stuff like this is often well-integrated in > Windows? Would you please check if there is newer BIOS? Oh, wait a second. So ACPI_C3 is C6, the deepest C-state one can request. Sorry, I missed this first. Scratch my questions about Windows and newer BIOS. So ACPI does expose the deepest C-state, but something prevents your system from going into PC10.
On Wed, 2020-08-26 at 19:25 +0300, Artem Bityutskiy wrote:
> C6
Ok, too long day, I meant C10 here...
On Wed, 2020-08-26 at 18:02 +0200, Rafael J. Wysocki wrote: > To that end, I would try to upgrade the graphics firmware and see if > you can get some nonzero PC8 residency then. I am curious, somehow that patch makes a difference. Guilhem, what do we actually compare: same kernel, just patched vs unpached? Or these are 2 different kernels, and one of them was patched. What does 'uname -r' say for the "with" and "without" kernels? Did you compile both kernels yourself and the only difference between them is the intel_idle patch?
On Wed, Aug 26, 2020 at 6:19 PM Artem Bityutskiy <dedekind1@gmail.com> wrote: > > Indeed, when I compare them: > > acpi_idle (without the patch): Does this come from the Guilhem's data? It's intel_idle in both cases, but in the "without the patch" case it uses ACPI. > CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 PkgWatt > 29.48 0.00 60.71 58 58 97.96 16.96 0.00 0.00 0.00 0.00 0.00 0.00 6.08 and I get the same data here, but > intel_idle (with the patch): > > CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 PkgWatt > 56 56 96.64 300 68.29 48.58 0.00 0.00 0.00 0.00 0.00 0.00 7.38 0.00 you seem to have columns wrong here. I get something like this CPU%c1 CPU%c6 CPU%c7 CoreTmp PkgTmp GFX%rc6 GFXMHz Totl%C0 Any%C0 GFX%C0 CPUGFX% Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 Pkg%pc8 Pkg%pc9 Pk%pc10 16.07 26.24 49.10 56 56 96.64 300 68.29 48.58 3.08 2.10 30.36 0.04 0.00 0.00 0.00 0.00 0.00 so still no PC10 residency (and it would be rather strange to get PC10 residency without any PC6 or higher residency). The 7.38 is the PkgWatt number AFAICS. > With intel_idle we reach PC10, without it we only go as deep as PC2 - huge difference. Not really. We don't get any PC10 residency in both cases. > I really wonder why the BIOS does not expose deeper C-states... It does expose C10. > And if it does not, is this for a reason? And how windows works then? It can only expose 3 C-states and it chose to expose C1, C7s and C10. > May be there is a BIOS update that fixes this problem? May be Windows > user get it quickly because stuff like this is often well-integrated in > Windows? Would you please check if there is newer BIOS? I doubt it. Cheers!
On Wed, Aug 26, 2020 at 6:39 PM Artem Bityutskiy <dedekind1@gmail.com> wrote: > > On Wed, 2020-08-26 at 18:02 +0200, Rafael J. Wysocki wrote: > > To that end, I would try to upgrade the graphics firmware and see if > > you can get some nonzero PC8 residency then. > > I am curious, somehow that patch makes a difference. It does make a difference, because it makes the processor spend more time in PC2. Which very well may be because the processor cannot enter deeper C-states.
I've done more tests, maybe it can give you more hints. I don't see that much differences between both (with and without patches) in this cases.
On Wed, 2020-08-26 at 18:43 +0200, Rafael J. Wysocki wrote: > > I am curious, somehow that patch makes a difference. > > It does make a difference, because it makes the processor spend more > time in PC2. Which very well may be because the processor cannot > enter deeper C-states. OK, you are right, we do not indeed have PC10 in both cases. The file was wrapped and I unwrapped it incorrectly. So indeed this is the real problem to solve, and the patch does not solve it.
On Wed, Aug 26, 2020 at 6:46 PM Guilhem Lettron <guilhem@barpilot.io> wrote: > > I've done more tests, maybe it can give you more hints. > I don't see that much differences between both (with and without > patches) in this cases. OK, thanks! I'm assuming that the topmost two sets of data are for the "without the patch" case whereas the other three correspond to the "with the patch" case. If so, the processor clearly enters PC10 in both cases and the residency percentages are similar. The numbers of times the POLL state was selected in the first test look kind of unusual (relatively very large), but other than this the patch doesn't seem to make much of a difference, so I'm not going to apply it. Thanks!
On Wed, 2020-08-26 at 19:00 +0200, Rafael J. Wysocki wrote: > On Wed, Aug 26, 2020 at 6:46 PM Guilhem Lettron <guilhem@barpilot.io> > wrote: > > > > I've done more tests, maybe it can give you more hints. > > I don't see that much differences between both (with and without > > patches) in this cases. > > OK, thanks! > > I'm assuming that the topmost two sets of data are for the "without > the patch" case whereas the other three correspond to the "with the > patch" case. I think the sample period is too short. Even with the same kernel, I can see the Busy% varies from 1% to 9%, and the PkgWatt varies from 0.4W to 2.4W. thanks, rui > > If so, the processor clearly enters PC10 in both cases and the > residency percentages are similar. > > The numbers of times the POLL state was selected in the first test > look kind of unusual (relatively very large), but other than this the > patch doesn't seem to make much of a difference, so I'm not going to > apply it. > > Thanks!
diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 8e0fb1a5bdbd..1bb539f09a4f 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -1145,6 +1145,8 @@ static const struct x86_cpu_id intel_idle_ids[] __initconst = { X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L, &idle_cpu_skl), X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE, &idle_cpu_skl), X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_X, &idle_cpu_skx), + X86_MATCH_INTEL_FAM6_MODEL(ICELAKE, &idle_cpu_skl), + X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_L, &idle_cpu_skl), X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X, &idle_cpu_icx), X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNL, &idle_cpu_knl), X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNM, &idle_cpu_knl),
Use the same C-states as SKL Signed-off-by: Guilhem Lettron <guilhem@barpilot.io> --- drivers/idle/intel_idle.c | 2 ++ 1 file changed, 2 insertions(+)